A Machine Learning-Optimized Immunogenic Cell Death Signature Reveals Tumor Immunogenicity and the Immunotherapy Response of Pancancer

Abstract

Tumor immunogenicity determines their response to immune checkpoint inhibitors (ICIs), but the mechanisms governing pancancer immunogenicity remain incompletely understood. A further critical barrier to developing reliable predictive biomarkers is data set shift, which undermines model generalizability. Here, we address these challenges by developing a novel adversarial validation (AV)-integrated machine learning framework, focusing on immunogenic cell death (ICD)-related gene signatures (ICDRSs). We designed three AV-based strategies to mitigate data set shift and validate the efficacies across multiple machine learning algorithms. Using dual-modal data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO), four optimal AV-based classifiers (e.g., GradientBoosting, XGBoost, LGBM, and CatBoost) were screened, which effectively reduced inter-cohort shift, enhancing both accuracy and robustness of downstream analysis. We identified novel risk/protective ICDRSs that strongly predicted patient survival and tumor immunogenicity across cancers. High-risk ICDRSs correlated with immune-exclusive microenvironments marked by impaired antigen presentation and aberrant tumor-associated macrophage development, as revealed by single-cell RNA sequencing. Validation across 13 ICI-treated cohorts revealed the capacity of ICDRSs for anti-PD-1 nonresponse. Mechanistically, risk ICDRSs were linked to CD47-SIRPA-mediated immune evasion and proliferative macrophage subsets with terminal dysfunction. This study advances understanding of tumor immunogenicity, provides novel biomarker development tools, and supports personalized cancer immunotherapy decision-making.

Keywords

adversarial validation / biomarker / immune checkpoint inhibitor / immunogenic cell death / immunogenicity / machine learning

References

[1] A. Ribas and J. D. Wolchok, “Cancer Immunotherapy Using Checkpoint Blockade,” Science 359, no. 6382 (March 2018): 1350-1355.

[2] T. E. Keenan, K. P. Burke, and E. M. Van Allen, “Genomic Correlates of Response to Immune Checkpoint Blockade,” Nature Medicine 25, no. 3 (March 2019): 389-402.

[3] P. M. K. Westcott, F. Muyas, H. Hauck, et al., “Mismatch Repair Deficiency Is Not Sufficient to Elicit Tumor Immunogenicity,” Nature Genetics 55, no. 10 (October 2023): 1686-1695.

[4] J. Sprooten, R. S. Laureano, I. Vanmeerbeek, et al., “Trial Watch: Chemotherapy-Induced Immunogenic Cell Death in Oncology,” Oncoimmunology 12, no. 1 (2023): 2219591.

[5] E. Catanzaro, R. Demuynck, F. Naessens, L. Galluzzi, and D. V. Krysko, “Immunogenicity of Ferroptosis in Cancer: A Matter of Context?,” Trends Cancer 10, no. 5 (May 2024): 407-416.

[6] S. T. Workenhe, J. Pol, and G. Kroemer, “Tumor-Intrinsic Determinants of Immunogenic Cell Death Modalities,” Oncoimmunology 10, no. 1 (March 2021): 1893466.

[7] M. Obeid, A. Tesniere, F. Ghiringhelli, et al., “Calreticulin Exposure Dictates the Immunogenicity of Cancer Cell Death,” Nature Medicine 13, no. 1 (January 2007): 54-61.

[8] T. Yamazaki, A. Kirchmair, A. Sato, et al., “Mitochondrial DNA Drives Abscopal Responses to Radiation That Are Inhibited by Autophagy,” Nature Immunology 21, no. 10 (October 2020): 1160-1171.

[9] G. Kroemer, C. Galassi, L. Zitvogel, and L. Galluzzi, “Immunogenic Cell Stress and Death,” Nature Immunology 23, no. 4 (April 2022): 487-500.

[10] T. Zhang, C. Yin, A. Fedorov, et al., “ADAR1 Masks the Cancer Immunotherapeutic Promise of ZBP1-Driven Necroptosis,” Nature 606, no. 7914 (June 2022): 594-602, https://doi.org/10.1038/s41586-022-04753-.

[11] P. Meier, A. J. Legrand, D. Adam, and J. Silke, “Immunogenic Cell Death in Cancer: Targeting Necroptosis to Induce Antitumour Immunity,” Nature Reviews Cancer 24 (March 2024): 299-315.

[12] M. W. Libbrecht and W. S. Noble, “Machine Learning Applications in Genetics and Genomics,” Nature Reviews Genetics 16, no. 6 (June 2015): 321-332.

[13] J. Montaner, L. Ramiro, A. Simats, et al., “Multilevel Omics for the Discovery of Biomarkers and Therapeutic Targets for Stroke,” Nature Reviews Neurology 16, no. 5 (May 2020): 247-264.

[14] X. Yang, L. Kui, M. Tang, et al., “High-Throughput Transcriptome Profiling in Drug and Biomarker Discovery,” Frontiers in Genetics 11 (2020): 19.

[15] J. Quionero-Candela, M. Sugiyama, A. Schwaighofer, and N. D. Lawrence, Dataset Shift in Machine Learning (MIT Press, 2009).

[16] J. G. Moreno-Torres, T. Raeder, R. Alaiz-Rodríguez, N. V. Chawla, and F. Herrera, “A Unifying View on Dataset Shift in Classification,” Pattern Recognition 45, no. 1 (2012): 521-530.

[17] A. Subbaswamy and S. Saria, “From Development to Deployment: Dataset Shift, Causality, and Shift-Stable Models in Health AI,” Biostatistics (Oxford, England) 21, no. 2 (April 2020): 345-352.

[18] S. G. Finlayson, A. Subbaswamy, K. Singh, et al., “The Clinician and Dataset Shift in Artificial Intelligence,” New England Journal of Medicine 385, no. 3 (June 2021): 283-286.

[19] Z. Zając, “Adversarial Validation, Part One,” Adversarial Validation, Part One Blog (May 2016), http://fastml.com/adversarial-validation-part-one/.

[20] H. Qian, B. Wang, P. Ma, L. Peng, S. Gao, and Y. Song, Managing Dataset Shift by Adversarial Validation for Credit Scoring (Springer Nature Switzerland, 2022), 477-488.

[21] J. Pan, V. Pham, M. Dorairaj, H. Chen, and J.-Y. Lee, Adversarial Validation Approach to Concept Drift Problem in Automated Machine Learning Systems. ArXiv. 2020;abs/2004.03045.

[22] Y. Ye, Q. Hu, H. Chen, et al., “Characterization of Hypoxia-Associated Molecular Features to Aid Hypoxia-Targeted Therapy,” Nature Metabolism 1, no. 4 (April 2019): 431-444.

[23] E. Azizi, A. J. Carr, G. Plitas, et al., “Single-Cell Map of Diverse Immune Phenotypes in the Breast Tumor Microenvironment,” Cell 174, no. 5 (August 2018): 1293-1308.e36.

[24] L. Tian, A. Goldstein, H. Wang, et al., “Mutual Regulation of Tumour Vessel Normalization and Immunostimulatory Reprogramming,” Nature 544, no. 7649 (April 2017): 250-254.

[25] W. Hugo, J. M. Zaretsky, L. Sun, et al., “Genomic and Transcriptomic Features of Response to Anti-PD-1 Therapy in Metastatic Melanoma,” Cell 165, no. 1 (March 2016): 35-44.

[26] H. Y. Chang, J. B. Sneddon, A. A. Alizadeh, et al., “Gene Expression Signature of Fibroblast Serum Response Predicts Human Cancer Progression: Similarities Between Tumors and Wounds,” PLoS Biology 2, no. 2 (February 2004): e7.

[27] Z. Hu, C. Fan, C. Livasy, et al., “A Compact VEGF Signature Associated With Distant Metastases and Poor Outcomes,” BMC Medicine 7 (March 2009): 9.

[28] P. Jiang, S. Gu, D. Pan, et al., “Signatures of T Cell Dysfunction and Exclusion Predict Cancer Immunotherapy Response,” Nature Medicine 24, no. 10 (October 2018): 1550-1558.

[29] D. A. Braun, Y. Hou, Z. Bakouny, et al., “Interplay of Somatic Alterations and Immune Infiltration Modulates Response to Pd-1 Blockade in Advanced Clear Cell Renal Cell Carcinoma,” Nature Medicine 26, no. 6 (June 2020): 909-918.

[30] M. Andreatta and S. J. Carmona, “UCell: Robust and Scalable Single-Cell Gene Signature Scoring,” Computational and Structural Biotechnology Journal 19 (2021): 3796-3798.

[31] L. Jerby-Arnon, P. Shah, M. S. Cuoco, et al., “A Cancer Cell Program Promotes T Cell Exclusion and Resistance to Checkpoint Blockade,” Cell 175, no. 4 (November 2018): 984-997.e24.

[32] Y. Mei, X. Wang, J. Zhang, et al., “Siglec-9 Acts as An Immune-Checkpoint Molecule on Macrophages in Glioblastoma, Restricting T-Cell Priming and Immunotherapy Response,” Nature Cancer 4, no. 9 (September 2023): 1273-1291.

[33] R. Bill, P. Wirapati, M. Messemaker, et al., “CXCL9:SPP1 Macrophage Polarity Identifies a Network of Cellular Programs That Control Human Cancers,” Science 381, no. 6657 (August 2023): 515-524.

[34] L. Garcia-Alonso, V. Lorenzi, C. I. Mazzeo, et al., “Single-Cell Roadmap of Human Gonadal Development,” Nature 607, no. 7919 (July 2022): 540-547.

[35] M. E. W. Logtenberg, F. A. Scheeren, and T. N. Schumacher, “The CD47-SIRPα Immune Checkpoint,” Immunity 52, no. 5 (May 2020): 742-752.

[36] M. A. Morrissey, N. Kern, and R. D. Vale, “CD47 Ligation Repositions the Inhibitory Receptor SIRPA to Suppress Integrin Activation and Phagocytosis,” Immunity 53, no. 2 (August 2020): 290-302.e6.

[37] S. Jhunjhunwala, C. Hammer, and L. Delamarre, “Antigen Presentation in Cancer: Insights Into Tumour Immunogenicity and Immune Evasion,” Nature Reviews Cancer 21, no. 5 (May 2021): 298-312.

[38] L. Galluzzi, I. Vitale, S. Warren, et al., “Consensus Guidelines for the Definition, Detection and Interpretation of Immunogenic Cell Death,” Journal for Immunotherapy of Cancer 8, no. 1 (March 2020): e000337.

[39] J. Huang, P. Chen, K. Liu, et al., “CDK1/2/5 Inhibition Overcomes Ifng-Mediated Adaptive Immune Resistance in Pancreatic Cancer,” Gut 70, no. 5 (May 2021): 890-899.

[40] B. Hu, C. Jin, H. B. Li, et al., “The DNA-Sensing AIM2 Inflammasome Controls Radiation-Induced Cell Death and Tissue Injury,” Science 354, no. 6313 (November 2016): 765-768.

[41] C. Wolf, E. L. Lim, M. Mokhtari, et al., “UNC93B1 Variants Underlie TLR7-Dependent Autoimmunity,” Science Immunology 9, no. 92 (February 2024): eadi9769.

[42] L. Bezu, A. Sauvat, J. Humeau, et al., “eIF2α Phosphorylation Is Pathognomonic for Immunogenic Cell Death,” Cell Death & Differentiation 25, no. 8 (August 2018): 1375-1393.

[43] M. J. Goldman, B. Craft, M. Hastie, et al., “Visualizing and Interpreting Cancer Genomics Data via the Xena Platform,” Nature Biotechnology 38, no. 6 (June 2020): 675-678.

[44] J. T. Leek, W. E. Johnson, H. S. Parker, A. E. Jaffe, and J. D. Storey, “The Sva Package for Removing Batch Effects and Other Unwanted Variation in High-Throughput Experiments,” Bioinformatics 28, no. 6 (March 2012): 882-883.

[45] R. A. Irizarry, “Exploration, Normalization, and Summaries of High Density Oligonucleotide Array Probe Level Data,” Biostatistics 4, no. 2 (April 2003): 249-264.

[46] A. Kauffmann, R. Gentleman, and W. Huber, “Arrayqualitymetrics--A Bioconductor Package for Quality Assessment of Microarray Data,” Bioinformatics 25, no. 3 (February 2009): 415-416.

[47] J. H. Wang, L. F. Zhao, H. F. Wang, et al., “GenCLip 3: Mining Human Genes’ Functions and Regulatory Networks From PubMed Based on Co-Occurrences and Natural Language Processing,” Bioinformatics 36, no. 6 (March 2020): 1973-1975.

[48] T. Akiba, S. Sano, T. Yanase, T. Ohta, and M. Koyama, “Optuna: A Next-Generation Hyperparameter Optimization Framework,” Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (July 2019): 2623-2631.

[49] S. M. Lundberg, G. Erion, H. Chen, et al., “From Local Explanations to Global Understanding With Explainable AI for Trees,” Nature Machine Intelligence 2, no. 1 (January 2020): 56-67.

[50] M. Sipper, “High Per Parameter: A Large-Scale Study of Hyperparameter Tuning for Machine Learning Algorithms,” Algorithms 15 (2022): 315.

[51] A. M. Newman, C. L. Liu, M. R. Green, et al., “Robust Enumeration of Cell Subsets From Tissue Expression Profiles,” Nature Methods 12, no. 5 (May 2015): 453-457.

[52] E. Becht, N. A. Giraldo, L. Lacroix, et al., “Estimating the Population Abundance of Tissue-Infiltrating Immune and Stromal Cell Populations Using Gene Expression,” Genome Biology 17, no. 1 (October 2016): 218.

[53] J. Racle, K. de Jonge, P. Baumgaertner, D. E. Speiser, and D. Gfeller, “Simultaneous Enumeration of Cancer and Immune Cell Types From Bulk Tumor Gene Expression Data,” eLife 6 (November 2017): e26476.

[54] X. Qiu, A. Hill, J. Packer, D. Lin, Y. A. Ma, and C. Trapnell, “Single-Cell Mrna Quantification and Differential Analysis With Census,” Nature Methods 14, no. 3 (March 2017): 309-315.

[55] M. Setty, V. Kiseliovas, J. Levine, A. Gayoso, L. Mazutis, and D. Pe'er, “Characterization of Cell Fate Probabilities in Single-Cell Data With Palantir,” Nature Biotechnology 37, no. 4 (April 2019): 451-460.

[56] H. Jung, H. S. Kim, J. Y. Kim, et al., “DNA Methylation Loss Promotes Immune Evasion of Tumours With High Mutation and Copy Number Load,” Nature Communications 10, no. 1 (Setember 2019): 4278.

RIGHTS & PERMISSIONS

2025 The Author(s). MedComm - Future Medicine published by John Wiley & Sons Australia, Ltd on behalf of Sichuan International Medical Exchange & Promotion Association (SCIMEA).